jvelja
/

prover_full_round_4_final_opt-state

Model card Files Files and versions

prover_full_round_4_final_opt-state / final

37 GB

1 contributor

History: 1 commit

jvelja's picture

🚀 End of training – full checkpoint with optimizer states

d9b755c verified 6 months ago

bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
Detected Pickle imports (8)
- "__builtin__.getattr",
- "torch._utils._rebuild_tensor_v2",
- "torch.FloatStorage",
- "collections.OrderedDict",
- "deepspeed.runtime.fp16.loss_scaler.LossScaler",
- "deepspeed.runtime.zero.config.ZeroStageEnum",
- "torch._tensor._rebuild_from_type_v2",
- "torch.Tensor"
How to fix it?
12.3 GB
xet

🚀 End of training – full checkpoint with optimizer states 6 months ago
bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
Detected Pickle imports (8)
- "torch._utils._rebuild_tensor_v2",
- "deepspeed.runtime.zero.config.ZeroStageEnum",
- "torch._tensor._rebuild_from_type_v2",
- "__builtin__.getattr",
- "torch.Tensor",
- "collections.OrderedDict",
- "deepspeed.runtime.fp16.loss_scaler.LossScaler",
- "torch.FloatStorage"
How to fix it?
12.3 GB
xet

🚀 End of training – full checkpoint with optimizer states 6 months ago
bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
Detected Pickle imports (8)
- "torch._utils._rebuild_tensor_v2",
- "torch.FloatStorage",
- "collections.OrderedDict",
- "torch.Tensor",
- "__builtin__.getattr",
- "torch._tensor._rebuild_from_type_v2",
- "deepspeed.runtime.fp16.loss_scaler.LossScaler",
- "deepspeed.runtime.zero.config.ZeroStageEnum"
How to fix it?
12.3 GB
xet

🚀 End of training – full checkpoint with optimizer states 6 months ago
zero_pp_rank_0_mp_rank_00_model_states.pt
Detected Pickle imports (5)
- "collections.OrderedDict",
- "torch._utils._rebuild_tensor_v2",
- "torch.BFloat16Storage",
- "__builtin__.set",
- "torch.Size"
How to fix it?
212 kB
xet

🚀 End of training – full checkpoint with optimizer states 6 months ago
zero_pp_rank_1_mp_rank_00_model_states.pt
Detected Pickle imports (5)
- "collections.OrderedDict",
- "torch._utils._rebuild_tensor_v2",
- "torch.BFloat16Storage",
- "__builtin__.set",
- "torch.Size"
How to fix it?
211 kB
xet

🚀 End of training – full checkpoint with optimizer states 6 months ago
zero_pp_rank_2_mp_rank_00_model_states.pt
Detected Pickle imports (5)
- "collections.OrderedDict",
- "torch._utils._rebuild_tensor_v2",
- "torch.BFloat16Storage",
- "__builtin__.set",
- "torch.Size"
How to fix it?
211 kB
xet

🚀 End of training – full checkpoint with optimizer states 6 months ago